search: improve with CompositeSuggestQueryParser

inveniosoftware · Dec 9, 2024 · ee9e9e7 · ee9e9e7
1 parent 11fe024
commit ee9e9e7
Show file tree

Hide file tree

Showing 2 changed files with 6 additions and 8 deletions.
diff --git a/invenio_users_resources/services/users/config.py b/invenio_users_resources/services/users/config.py
@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 # Copyright (C) 2022 TU Wien.
-# Copyright (C) 2022 CERN.
+# Copyright (C) 2022-2024 CERN.
 # Copyright (C) 2024 KTH Royal Institute of Technology.
 #
 # Invenio-Users-Resources is free software; you can redistribute it and/or
@@ -26,10 +26,10 @@
     SortParam,
 )
 from invenio_records_resources.services.records.queryparser import (
+    CompositeSuggestQueryParser,
     FieldValueMapper,
     QueryParser,
     SearchFieldTransformer,
-    SuggestQueryParser,
 )
 from luqum.tree import Word
 
@@ -68,9 +68,9 @@ class UserSearchOptions(SearchOptions, SearchOptionsMixin):
     # The user search needs to be highly restricted to avoid leaking
     # account information, hence do not edit here unless you are
     # absolutely sure what you're doing.
-    suggest_parser_cls = SuggestQueryParser.factory(
+    suggest_parser_cls = CompositeSuggestQueryParser.factory(
         tree_transformer_cls=SearchFieldTransformer,
-        fields=["username^2", "email^2", "profile.full_name^3", "profile.affiliations"],
+        fields=["username^2", "email.keyword^2", "profile.full_name^3", "profile.affiliations"],
         # Only public emails because hidden emails are stored in email_hidden field.
         allow_list=["username", "email"],
         mapping={
@@ -81,7 +81,6 @@ class UserSearchOptions(SearchOptions, SearchOptionsMixin):
             "name": "profile.full_name",
         },
         type="most_fields",  # https://www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl-multi-match-query.html#multi-match-types
-        fuzziness="AUTO",  # https://www.elastic.co/guide/en/elasticsearch/reference/current/common-options.html#fuzziness
     )
 
     params_interpreters_cls = [

diff --git a/tests/services/users/test_service_users.py b/tests/services/users/test_service_users.py
@@ -111,10 +111,9 @@ def test_user_search_field_not_searchable(user_service, user_pub, query):
         ("CERN", USERNAME_BOTH),
         ("Jose", USERNAME_JOSE),
         ("Jos", USERNAME_JOSE),
-        ("Jose CERN", USERNAME_BOTH),
-        ("Jose AND CERN", USERNAME_BOTH),
+        ("Jose CERN", USERNAME_JOSE),
         ("Tim", USERNAME_TIM),
-        ("Tim CERN", USERNAME_BOTH),
+        ("Tim CERN", USERNAME_TIM),
         ("[email protected]", USERNAME_JOSE),
         ("[email protected]", USERNAME_JOSE),
         ("pub@inveniosoft", USERNAME_JOSE),